The primary objective of this proposal is to complete the data analysis for the Cholesterol Lowering Atherosclerosis Study (CLAS) -- a randomized, placebo controlled, angiographic clinical trial providing information on human atherosclerosis regression/progression in coronary, femoral, and carotid arteries. Two and four year endpoint data have been collected for coronary (panel reading and image processed), femoral (image processed) and carotid arteries (image processed). Risk factor data are available on baseline and on-trial clinical measures, lipids, apolipoproteins and nutrient intake. Unique features of CLAS include: 1) simultaneous angiography and image analysis of carotid, femoral and coronary arteries; 2) the opportunity to observe arterial wall risk factor effects of a wide range of plasma cholesterol levels (185 to 350 mg%) without a confounding influence from smoking; and 3) the opportunity to evaluate dietary effects on atherosclerosis regression/progression in an extensive nutritional database. Statistical evaluation will include both a per-segment analysis (taking into account segment inter-dependence) and a per-patient analysis. These analyses will address an important conceptual problem inherent in vessel image based trials, the issue of the independence of segments/lesions within a patient.